engaged and productive misfits: how job crafting...

r Academy of Management Journal2016, Vol. 59, No. 5, 1561–1584.http://dx.doi.org/10.5465/amj.2014.0850

ENGAGED AND PRODUCTIVE MISFITS: HOW JOBCRAFTING AND LEISURE ACTIVITY MITIGATE THENEGATIVE EFFECTS OF VALUE INCONGRUENCE

RYAN M. VOGELPennsylvania State University–Erie

JESSICA B. RODELLUniversity of Georgia

JOHN W. LYNCHUniversity of Illinois at Chicago

The work life of “misfits”—employees whose important values are incongruent withthe values of their organization—represents an under-researched area of the person–environment fit literature. The unfortunate reality is that these individuals are likely to bedisengaged and unproductive at work. In thismanuscript, we entertain the possibility thatemployees can protect themselves from this situation if they engage in alternative actionsthat supplement the fundamental needs that go unmet from value incongruence. Weintegrate theorizing about the motivational role of need fulfillment and work/nonworkbehaviors in order to examine whether two actions in particular—job crafting and leisureactivity—can potentially mitigate the negative effects of value incongruence on employeeperformance. The results from our field study of employees from diverse organizationsand industries suggest that both job crafting and leisure activity do indeed act as a buffer,mitigating the otherwise negative effects of value incongruence on employee engagementand job performance (with regard to both task performance and citizenship behavior).

The topic of “value congruence”—the alignment ofemployee and company values—poses an interestingdilemma for managers in today’s workplace. In bothresearchandpractice, therehasbeenanoverwhelmingfocus on the importance and benefits of “fit” betweenan employee’s values and the company’s values(Kristof-Brown & Guay, 2011). Such alignment is be-lieved to help strengthen organizational culture, im-prove efficiency, and retain employees (e.g., Edwards& Cable, 2009; Meglino, Ravlin, & Adkins, 1989). De-spite the benefits of alignment, value incongruencemayalso servean important function in theworkplace.“Misfit”—defined here as when an employee’s im-portant values are incongruent with the values of theirorganization (Kristof, 1996)—presents an opportunityfor increased diversity in priorities andperspectives toemergewithinanorganization.Suchdiversitycan leadto increased creativity (e.g., Hoever, van Knippenberg,vanGinkel, & Barkema, 2012). Similarly, some level ofmisfit can help organizations avoid rigidity and stag-nation, which detract from organizational effective-ness (Harrison, 2007; Schneider, 1987).

From the perspective of the employee, however,the experience of misfit has been associated with

uniformly negative outcomes. The misalignment ofimportant values can leave misfits with a sense thatthey do not belong and of feeling unfulfilled by theirwork (Edwards & Cable, 2009; Edwards & Shipp,2007; Rich, LePine, & Crawford, 2010). For example,an individual who strongly values personal statusand being looked up to by others would see littlemeaning in—and feel out of place working for—anorganization inwhich the predominant value systememphasizes teamwork and interdependence. Asa result, the experience of misfit can be stressful,uncomfortable, and associated with negative workattitudes and behaviors. Indeed, misfits have greaterintentions to quit, are more likely to turn over, andare likely to experience lower job satisfaction, en-gagement, and performance (Edwards & Cable, 2009;Kristof-Brown, Zimmerman, & Johnson, 2005; Richet al., 2010; Schneider, 1987).

Moreover, the nature of today’s business environ-ment has increased the prevalence of the misfitphenomenon (Moore, 2010; Sthapit, 2010). Recenttrends in recruiting and hiring approaches—such as“selling” applicants on the job and giving shortwindows for applicants to decide on job offers—have

1561

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

http://dx.doi.org/10.5465/amj.2014.0850

resulted in more misfits working in organizations(Sthapit, 2010). This situation is only predicted toworsen in the coming years, as the values of youn-ger employees and companies are particularly mis-matched (Moore, 2010). Because individual valuesare relatively stable over time (Rokeach, 1973), em-ployees hired with initial value incongruence arelikely to continue to experience it throughout theirtenure with an organization. Although theorizingwould predict that misfits ultimately quit their jobs(Schneider, 1987), poor labormarket conditionsmayimpede this otherwise-natural correction. Recentdata indicate that a recuperating yet weak labormarket has leftmanymisfits unable to find alternativeemployment despite their desire to move (Harding &Mackenzie, 2014; Light, 2010; U.S. Bureau of LaborStatistics, 2014).

Given this state of affairs—the prevalence of mis-fits combined with mixed implications for compa-nies and employees—we believe that the topic ofmisfit deserves more specific research attention.Specifically, we would like to consider the pos-sibility that not all misfits are destined to suffera disengaging and unproductive work experience.Meta-analytic evidence has shown that there is con-siderable variability in the effects of value congruenceon performance (Kristof-Brown et al., 2005), suggest-ing theremaybe factors that offset thenegative impactof value incongruence. Uncovering these factors maynot only demonstrate how employees can survive ina company as misfits, but also how companies maybenefit from having different values among their em-ployees while minimizing the performance detri-ments of value incongruence.

In particular, we suggest that employeesmay havesome direct control over whether or not their expe-rience of misfit is detrimental to their job perfor-mance. Recent trends in the management literaturehighlight employees’ ability and desire to proactivelymanage their life experiences (see Grant & Ashford,2008, for a review).These scholars are acknowledgingthat employees do not simply react to the worldaround them, but, rather, they behave in ways thatcontribute to shaping their environment (Grant &Ashford, 2008; Parker, Bindl, & Strauss, 2010). In-tegrating this perspective with the traditional view ofmisfit introduces the possibility that employees maybe able to proactively combat the otherwise detri-mental effects of misfit. In a way, then, employeesmay “co-create” their environment, providing them-selves with a buffer from the experience of value in-congruence, and ultimately improving their situationat work.

Adopting this perspective, our goal is to examinethe following question: What proactive behaviorsmight buffer the performance detriments of valueincongruence? To explore potential answers to this,we draw from the literature on work/nonwork do-mains (Edwards & Rothbard, 2000; Greenhaus &Powell, 2006). Although the majority of research onproactive employee behavior has focused on thework domain (Grant & Ashford, 2008; Parker et al.,2010), there is increasing acknowledgment that be-havior outside of the workplace can also have a signif-icant impact on employees’ attitudes and experiencesat work (Edwards & Rothbard, 2000; Greenhaus &Powell, 2006).Thus,wewill consider the effectivenessof proactivity in both work and nonwork as a way ofsupplementing the core needs that might be unmet formisfits.

One of the most direct forms of proactivity atwork is job crafting, which refers to an employee’svolitional actions to shape,mold, and redefine theirjob in an attempt to improve their experience ofwork (Wrzesniewski & Dutton, 2001; Wrzesniewski,LoBuglio, Dutton, & Berg, 2013). Proactivity in thenonwork domain can be seen in the choices thatpeople make regarding how to use their time; to theextent that they volitionally engage in activities thatcan be fulfilling, they are behaving more proactively.Thus, we also examine employees’ leisure activity—defined as the extent to which employees engage innonwork activities, such as exercising, reading, orother hobbies (Gowen, Riordan, & Gatewood, 1999).

Thus, the purpose of this manuscript is to explorethe potential of job crafting and leisure activity to actas buffers formisfits. Themodel, depicted inFigure 1,is grounded by coupling motivation theorizing aboutneed fulfillment (e.g., Baumeister & Leary, 1995; Deci&Ryan, 1985, 2000;Kahn, 1990, 1992)with thework/nonwork literature (e.g., Edwards & Rothbard, 2000;Greenhaus & Powell, 2006). The underlying logicfrom this theorizing is that job crafting and leisureactivity should act as resources that satisfy employeeneeds that would otherwise be deprived from expe-riencing value incongruence. Thus, job crafting andleisure activity have the potential to offset the nega-tive motivational effects of value incongruence onemployee performance (as regards both task perfor-mance and citizenship behavior).

Fit scholars tend to direct their focus toward thepsychological experience of value congruence, at theexpense of specific examinations of value incon-gruence. Indeed, “misfit is a subject that has beenlargely overlooked by researchers . . . we know littleabout how they behave or cope” (Kristof-Brown &

1562 OctoberAcademy of Management Journal

Guay, 2011: 38–39). By focusing on this experienceof misfit, our manuscript offers theoretical contri-butions to the value congruence literature and, morebroadly, to the person–environment (P–E) fit litera-ture. For example, this paper refines and broadensour understanding of one of the dominant theoreticalframeworks in this area: the attraction–selection–attrition (ASA)model (Schneider, 1987).ASA theoryproposes that value incongruence is likely to promptmisfits to experience negative work attitudes, and,consequently, to leave their organizations. Our studyextends this view by painting employees in a moreproactive light, wherein misfits can combat thisnegative situation—essentially “co-creating” theirwork experience so that it is more fulfilling and en-gaging for them, despite value incongruence.

Our research further contributes to the value con-gruence and P–E fit literatures by examining bothwork (job crafting) and nonwork (leisure activity)behaviors as ways to compensate for value incon-gruence. This manuscript is the first that we knowof to incorporate a cross-domain perspective byconsidering the buffering effects of leisure activity.Although P–E fit scholars have mainly focused onwithin-domainrelationships (e.g.,Edwards&Rothbard,1999), recent research has demonstrated the value ofconsidering how nonwork endeavors can impact workexperiences (e.g., Trenberth & Dewe, 2005). Moreover,leisure activity has become more important witheach generation of employees (Twenge, Campbell,

Hoffman, & Lance, 2010), pointing to a need to betterunderstand how itmight impact experiences at work.

TRADITIONAL PERSPECTIVE OF VALUEINCONGRUENCE

“Values,”definedas fundamental and stablebeliefsabout preferred end states or behaviors (Rokeach,1973), describe how individuals see themselves attheir core, thereby guiding their judgments andchoices, influencing their attitudes, and shaping theirbehaviors (Meglino & Ravlin, 1998). A situation of“value congruence”—in which individual valuesmatchorganizational values—is generally consideredideal for employees (Kristof-Brown et al., 2005). Forexample, an employee who highly values autonomyis more likely to thrive in an organization that priori-tizes freedom than in an organization characterizedby micromanagement. The underlying reasoning isthat a match between an individual and his/her workenvironment satisfies one or more of a person’s basicpsychological needs (Baumeister & Leary, 1995; Deci& Ryan, 1985; Edwards & Shipp, 2007; Heine, Proulx,& Vohs, 2006; Kahn, 1990; Ryan & Deci, 2000). Theseneeds can range from a desire to be connected withothers to having freedom over one’s own actions tolonging for a senseofpurposeormeaning (Baumeister& Vohs, 2005; Deci & Ryan, 1985; Ryan & Deci, 2000;Williams, 1997). Such needs are so innate and es-sential that people are not necessarily directly aware

FIGURE 1Theoretical Model

2016 1563Vogel, Rodell, and Lynch

of the particular needs that are driving their goals orbehaviors (Deci & Ryan, 2000). Regardless, fulfillingthese needs can have a vast and profound impact onemployee attitudes and behaviors.

Value incongruence (or “misfit”), however, is gen-erally considered to be detrimental for employees(e.g., Kristof-Brown et al., 2005; O’Reilly, Chatman, &Caldwell, 1991). As employee values diverge fromorganizational values, their core needs become de-prived,which is in turnharmful toemployeeattitudesand behaviors (Baumeister & Leary, 1995; Deci &Ryan, 1985; Edwards&Shipp, 2007). It is important tonote that misfit may be detrimental regardless of thedirection it takes—that is, whether an employeevalues something more or less than their company(Cable & Edwards, 2004). Consider, for example,a situation of mismatch regarding salary. Value con-gruence is not an issue of whether an employee issatisfied with his/her current level of pay.1 Rather, itcaptures whether the employee feels that he or sheprioritizes pay issues to the same degree as theorganization.

Kahn’s (1990, 1992) seminal theorizing speaks tothe motivational implications of value congruence(and, thus, incongruence). Employees experiencingvalue congruence see their true self as being alignedwith their company’s expectations of them (Chatman,1989; Edwards&Shipp, 2007;Kahn, 1990, 1992). Thisalignment reinforces an individual’s preferred self-image,making itmore likely that they view their job ascompellingandworthwhile (Kahn,1992)—somethingthat can fulfill a variety of psychological needs(Baumeister & Vohs, 2005; Deci & Ryan, 1985; Ryan&Deci, 2000;Williams, 1997). Indeed, Greguras andDiefendorff (2009) found that perceived value con-gruence enabled employees to feel that theybelonged,werecompetent, andhaddiscretionover their actions.

Because it satisfies core needs, value congruence isconsidered intrinsically motivating (Ryan & Deci,2000). In particular, it should foster job engagement—wherein employees invest their whole selves (cogni-tively, behaviorally, and emotionally) into theirjobs—which has implications for their job performance(Kahn, 1990, 1992; Rich et al., 2010). Employees whofully invest themselves into their work should exhibitstronger task performance and citizenship behaviorbecause such a holistic investment facilitates persis-tence to goal accomplishment, increased attention to

work tasks, and greater cooperation with others (Kahn,1990). In support of this theorizing, Rich et al. (2010)found that perceived value congruence increased jobengagement and, ultimately, task performance and cit-izenship behavior.

At the opposite end of this experience are em-ployees whose values do not fit with their organiza-tion’s values. These employees perceive that they areexpected to act like someone they are not (Kahn,1990, 1992). They are likely to sense that they arelacking something important and valuable tothem—that a fundamental need is not quite fulfilled(Edwards & Shipp, 2007). For example, misfits arelikely to have fewer and less enjoyable interpersonalwork relationships than those who fit (Edwards &Cable, 2009), leaving belongingness needs un-fulfilled (Baumeister & Leary, 1995; Edwards &Shipp, 2007). Because misfits’ time is spent pursu-ing goals contrary to their ideal self and core values,they are also likely to believe that their work lacksmeaning (Kahn, 1990, 1992). As a result, misfitsshould be less willing to fully engage and investthemselves into their work, causing their job per-formance to suffer (Kahn, 1990, 1992).

A BUFFERING PERSPECTIVE ON VALUEINCONGRUENCE

Despite the ability of value congruence to fulfillemployees’ needs, there are also other sources ofpotential need fulfillment in an individual’s life thatcould contribute to well-being, engagement, andquality performance (Deci &Ryan, 1985;Kahn, 1990,1992; Ryan & Deci, 2000). Indeed, employees pro-actively engage in all sorts of actions—both withinthe work domain and outside of it—that present theopportunity to satisfy their basic needs. Thus, theperformance implications of misfit may not be assimple and straightforward as traditionally assumed.Instead, to create a more comprehensive under-standing of the implications ofmisfit, we should alsotake into account these other sources of need fulfill-ment. In this section, we theorize about the potentialbuffering role of two forms of proactive behavior—job crafting and leisure activity—that may providethis opportunity for employees.

The satisfaction of needs is subject to the princi-ples of substitution and satiation (Baumeister &Leary, 1995; Heine et al., 2006). These principlessuggest that (a) if someone’s needs are not providedby a particular source, they may be filled by othermeans, and (b) if a need is sufficiently satisfied, otherpotential sources should offer diminishing returns.

1 That issue would be captured more clearly in what isreferred to as “needs–supplies fit,”which could be used asan indicator of whether the amount of money needed bythe employee is actually provided by the organization.


Baumeister (2012) illustrated this process throughthe idea of hunger. He suggested that, althoughhunger could be brought on by thoughts of a ham-burger, this need could just as easily be satisfiedwitha sandwich or plate of spaghetti and not necessarilyahamburger. In addition, this sandwichwould likelybe less satisfying to a person who has just consumeda large meal than to one who has not eaten in severalhours.

Applied to the current context, this theorizingsuggests that other sources of need satisfaction—such as job crafting (Berg, Grant, & Johnson, 2010;Rich et al., 2010) and leisure activity (Clary et al.,1998; Gowen et al., 1999)—may serve as a substitutefor the needs lacking due to value incongruence.These activities, however, should be less relevant toemployees who have achieved a better fit with thevalues of their organizations because their needs arealready fulfilled to a greater extent (i.e., satiated). Inthe following sections, we describe how the pro-active nature of job crafting and leisure activity canfulfill employees’ basic needs and, thus, may bufferthe de-motivational consequences of misfit on jobperformance.

Job Crafting as a Buffer of Misfit

The goal of job crafting is to change the parame-ters of one’s job to suit personal needs, preferences,and abilities (Berg, Dutton, & Wrzesniewski, 2013).Wrzesniewski and her colleagues have stressed thatany employee, even thosewithout formal discretionover their role, can craft their job in significant ways(Wrzesniewski & Dutton, 2001;Wrzesniewski et al.,2013). Job crafting occurs in two general forms:changes to either (or both) the design of one’s joband the social environment at work (Wrzesniewski& Dutton, 2001). Changes in job design include in-troducing new approaches to help improve one’swork, adding preferred tasks or minimizing un-enjoyable tasks, and changing standard proceduresof the job (e.g., Leana, Appelbaum, & Shevchuk, 2009).Such changes can satisfy employees’ core needs. Forexample, changes to the job can enhance meaning anemployee obtains from their job, allow them to feelmore competent at their job, or provide them withmore autonomy in their job (Hackman & Oldham,1980; Kahn, 1990, 1992).

Crafting the social environment at work includeschanges to the quality or amount of interactions withothers or altering the nature of relationships in waysthat changeone’s job (Leanaetal., 2009;Wrzesniewski& Dutton, 2001). In particular, job crafting can offer

employees theopportunity to increase interactionswithand create stronger connections to colleagues or clientswhomtheyprefer, andminimize (or avoid) interactionsthat they expect to be unpleasant. Crafting the socialenvironment in thesewayscanprovideemployeeswithmore supportive and rewarding interactions, as well ashigher-qualitygroupdynamics—offering fulfillment foremployees in terms of, for example, meaning and be-longing (Baumeister & Leary, 1995; Berg et al., 2010;Kahn, 1990).

By substituting for needs unsatisfied due to valueincongruence in theseways, job crafting shouldbufferthe motivational consequences of misfit on job per-formance. According to Kahn’s (1990, 1992) theoriz-ing, the situations that result from jobcrafting—wheretask characteristics and the social environment aretailored to the employee’s own preferences—set thestage for greater jobengagement. Inparticular, craftingone’s job to better fulfill needs such as meaning, au-tonomy, and competence enhances employees’ be-liefs that they will personally benefit from investingthemselves in their work—something that Kahn(1990) considered a psychological condition neces-sary for greater engagement. Furthermore, envi-ronments characterized with high-quality workinteractions and relationships should provide em-ployees with a sense of safety and belonging—wherein they feel that they can display their trueself to others—which also fosters job engagement(Kahn, 1990). Thus, we expect that job crafting willhelp satisfy needs for misfits, thereby mitigating theotherwisenegative impactofvalue incongruenceon jobengagement. On the other end of the spectrum—whereemployees experience greater value congruence—weexpect that jobcraftingwillhave relatively little impact,as the needs of these employees are presumably metdue to being a good fit.

Hypothesis 1. The negative effect of value in-congruence on job engagement will be weakerwhen job crafting is high than when it is low.

Through this impact on employee engagement, jobcrafting should also buffer the negative implicationsof value incongruence for job performance, both interms of task performance and citizenship behavior.Task performance refers to in-role behaviors thatare directly related to a job’s core tasks (Borman &Motowidlo, 1993). Engaged employees should ex-hibit stronger task performance because they workharder and for longer periods of time, pay more at-tention to their tasks, and emotionally connect totheir role (Kahn, 1990; 1992; Rich et al., 2010). Citi-zenship behavior refers to discretionary behaviors


that, although not directly related to the core tasks ofone’s job, also contribute to an organization’s effec-tive functioning (Organ, 1988). In Kahn’s (1990)theorizing about the role of engagement, he proposedthat engaged employees would have a broader con-cept of their work role. In this way, engagement hasbeen considered to be an indicator of an employee’swillingness to go beyond their formal job de-scriptions and act in ways that help their employer(Christian, Garza, & Slaughter, 2011; Rich et al.,2010). In particular, we consider the impact of en-gagement on what Williams and Anderson (1991)termed citizenship behaviors directed at the organi-zation, given that this focus matches the organiza-tional target of value incongruence (Lavelle, Rupp, &Brockner, 2007). These behaviors include employeeactions such as speaking up to improve organiza-tional functioning and looking out for the company’sbest interests.

Given the proposed buffering effect of job craftingon job engagement, we argue that job crafting shouldultimately influence the indirect relationship be-tween value incongruence and job performance. Bysubstituting for value incongruence, job craftingmaybuffer employees from the otherwise negative im-pact on employees’ job performance through theirengagement (Baumeister & Leary, 1995; Kahn, 1990,1992; Wrzesniewski & Dutton, 2001). Despite a lackof fit for these individuals, theymaybeable to remainengaged and productive in their jobs because theyhave found alternative sources that are fulfilling andprovide the necessary physiological resources todo so.

Hypothesis 2. The negative indirect effect ofvalue incongruence on (a) taskperformance and(b) citizenship behavior (via job engagement)will be weaker when job crafting is high thanwhen it is low.

Leisure Activity as a Buffer of Misfit

An alternative option to buffer the negative con-sequences ofmisfit may lie outside of theworkplace.Researchers are increasingly acknowledging thatnonwork activities play a significant role in workexperiences (Greenhaus & Beutell, 1985; Greenhaus& Powell, 2006). Latack and Havlovic (1992) specu-lated that leisure activity—such as personal hobbies,exercise and sports, and community involvement—could provide employees with an opportunity todealwithnegative situations atwork (see alsoGowenet al., 1999). Because people have considerable

freedom to choose the type of leisure activities topursue, such activities present a unique opportunityfor people to fulfill their needs and desires. The in-herent discretion over leisure activity can provideindividuals with a sense of autonomy (Ryan & Deci,2000). In addition, peoplemay choose hobbies (suchas playing a musical instrument in a band) so thatthey can master something and demonstrate theircompetence, as well as to socialize and satisfy be-longingness needs (Deci & Ryan, 1985; Ryan & Deci,2000). There is also evidence that community in-volvement provides employees with greater mean-ing and connections to others (Clary et al., 1998;Geroy, Wright, & Jacoby, 2000).

By offering this sense of fulfillment, leisure activ-ity may serve as a substitute for value incongruence(Baumeister & Leary, 1995). In its most traditionalsense, motivation theorizing may imply that thisprocess would foster greater involvement in thatparticular leisure activity (Deci & Ryan, 1985; Kahn,1990, 1992; Ryan & Deci, 2000). Incorporating ideasfrom the work/nonwork literature, however, pres-ents the possibility that leisure activity may alsohave motivational implications for the workplace.General theorizing about nonwork activities (Burke& Greenglass, 1987; Edwards & Rothbard, 2000), aswell as more specific research on leisure activities,suggest that leisure activities can serve as a substitutefor unmet needs in the workplace (e.g., Grant, 2012;Heine et al., 2006; Tinsley & Eldredge, 1995; Tinsley,Hinson, Tinsley, & Holt, 1993).

Using Kahn’s (1990) terminology to describe thisprocess, it is possible that leisure activity createsa sense of “psychological availability” for job en-gagement. Kahn (1990, 1992) theorized that a state ofpsychological availability exists when employeeshave the physical, emotional, and physiological re-sources necessary to engage in their jobs. AlthoughKahn (1990) admitted that life outside ofwork—suchas leisure activity—could sometimes divert energythat could otherwise be devoted to work, he alsotheorized that such activity has the potential to in-crease energy and availability for work. Theorizingabout the synchronization of work and nonworkdomains echoes this suggestion—that nonwork ac-tivities can spill over to work and compensate fordeficiencies that people experience at work (Burke &Greenglass, 1987;Greenhaus&Powell, 2006; see alsoEdwards & Rothbard, 2000). Scholars have foundthat leisure activity creates positive emotional statesthat extend into the workplace (Hecht & Boies, 2009;Hills & Argyle, 1998). Moreover, there is evidencethat various forms of leisure activity serve as a form


of recovery for employees—renewing their energysupply and enabling them to remain focused atwork (Sonnentag, 2003; Sonnentag,Mojza, Demerouti,& Bakker, 2012; Trougakos, Beal, Green, & Weiss,2008).

Ultimately, then, we expect that leisure activitymay compensate for needs unsatisfied from valueincongruence—acting as a buffer from the otherwisedetrimental effects on job engagement and perfor-mance that befall misfits (Baumeister & Leary, 1995;Burke & Greenglass, 1987; Kahn, 1990, 1992). More-over, aswas the casewith job crafting, the principle ofsatiation should also be applicable to leisure activity.For employeeswith greater value congruence, leisureactivity will be less relevant and impactful becausethe needs of these employees are already satisfied toa greater extent (Baumeister & Leary, 1995).

Hypothesis 3. The negative effect of value in-congruence on job engagement will be weakerwhen leisure activity is high than when it is low.

Hypothesis 4. The negative indirect effect ofvalue incongruence on (a) taskperformance and(b) citizenship behavior (via job engagement)will be weaker when leisure activity is high thanwhen it is low.

METHOD

Participants and Procedures

Our study design included three time points andtwo sources. To achieve sufficient variance on bothindividual and organizational values, we recruitedparticipants through online classified advertise-ments, similar to other studies in top managementjournals (e.g., Colquitt, Baer, Long, & Halvorsen-Ganepola, 2014; Judge, Simon,Hurst, &Kelley, 2013;Rodell & Judge, 2009). To qualify for our study, par-ticipants had to be at least 18 years old, work full-time (at least 35 hours per week), and provide thename, email address, and mailing address for a su-pervisor who could complete one survey on theirbehalf. Upon responding to the advertisement, par-ticipants received their first survey (time 1), whichmeasured value congruence, job crafting, and leisureactivity, as well as general personality and de-mographics. Three weeks later, we emailed partici-pants the second survey (time 2), which measuredjob engagement. Three weeks after the second sur-vey, we emailed supervisors their survey (time 3),which included measures of the participants’ job per-formance (taskperformanceandcitizenshipbehavior).

Several steps were taken to ensure the legitimacy ofsupervisors. For example, we removed potential par-ticipants who either provided their own name and/oremail address instead of a supervisor’s name and/oremail address, or “suspicious” email addresses (i.e., itwas clearly the participant’s personal address). Fur-ther, we carefully checked whether participants’ andsupervisors’ physical mailing addresses were identi-cal; when they were, we contacted the participant di-rectly to determine whether there was a legitimateexplanation (e.g., bothwere thephysical address of theorganization). We also inspected email and physicaladdresses for participants attempting to complete thestudy more than once. For their time, participants re-ceived $5, and both participants and supervisors wereentered into a draw for $500.

Initially, 404 participants expressed interest in thestudy by completing the first survey. Of these, 302completed the second survey, and 206 supervisorsresponded to their survey. To bolster response rates,we sent up to four reminders to participants and/ortheir supervisors. After matching responses andaddressing missing data and outliers, the final data-set included 193 respondents, representing a 47.8%participation rate. Of the participants, 55.0% werefemale, 61.5% were Caucasian, the average age was34.96 years (SD 5 10.49), and the average organiza-tional tenure was 5.65 years (SD 5 4.73). Of the su-pervisors, 46.1%were female, 64.1%wereCaucasian,the average age was 41.85 years (SD5 11.49), and theaverage organizational tenure was 8.94 years (SD 56.23). The sample represented many industries; thefour most common were finance and banking, in-formation technology, education, and health care.

Measures

Value incongruence. Value incongruence wasmeasured using the 24 items of the Work ValuesSurvey from Cable and Edwards (2004). The WorkValues Survey is based on the circumplex model ofhuman values developed by Schwartz (1992) and isdesigned to assess eight core individual and organi-zational work values: altruism, relationships, pay,security, authority, prestige, variety, and autonomy.FollowingCable and Edwards (2004), participants inour study assessed both their personal values—byanswering the question “How important is this toyou?”—and their organization’s values—by answer-ing the question “How important is this to yourcompany?”—with respect to each of the 24 items (3items for eachworkvalue). Participantsuseda5-pointscale ranging from 1 (Not at all important) to


5 (Extremely important). Coefficient alphas for in-dividuals’ values ranged from .80 to .91; for the or-ganizations’ values, they ranged from .83 to .93.

Not all values are equally important to each em-ployee, and employees likely do not prioritize thesame values as one another (Rokeach, 1973). Theextent to which a particular value is relevant to anindividual’s job attitudes and behaviors (such asengagement and performance) is likely dependenton the relative importance of that value to that per-son. Indeed, researchers have acknowledged thatmisfit is likely to bemore detrimental when it occurswith more strongly held values (Edwards, 1992;Edwards, 1996; Locke, 1976). Thus, we could rea-sonably expect that the value considered most im-portant to an individual plays the most significantrole in explaining job attitudes and behaviors, andthat there would be diminishing explanatory powerfrom the values deemed less important. Because thisstudy centers around thepotential outcomes of valueincongruence—and not on the differences betweenvalue congruence for the eight values—we focusedour analyses on the most important value to eachparticipant.

To determine the most important value, we askedeach participant to rank order the eight core values.Participants were provided with definitions of thesevalues that were drawn from the organizational lit-erature (e.g., “Altruism: doing things to help otherswithout payment or material reinforcement”) andranked the relative importance of each dimensionfrom 1 (Most important) to 8 (Least important).Rather than simply using the highest-rated value foreach participant, which could result in equal ratingsfor multiple values, asking participants to rank orderthe values allowed us to most clearly determine theone value considered most important to each indi-vidual. Indeed, in 71.5% of the cases, the rank or-dering approach was necessary to determine theindividual’s most important value. Value incongru-ence was operationalized using the three items rep-resenting the rating for the individual andorganizationon theparticular value rankedasmost important to theindividual.

Job crafting. Job crafting was measured withLeana et al.’s (2009) 4-item measure. On a 5-pointscale ranging from1 (Almost never) to 5 (Very often),participants were asked to rate how often they en-gaged in each of the listed behaviors. Sample itemsincluded “Introduce new approaches to improvemy work” and “Change minor work proceduresthat I think are not productive.” The coefficient awas .83.

Leisure activity. Participants were shown a list ofcommon leisure activities (e.g., exercising, enter-tainment, other hobbies) and asked to rate theiragreement with each of 7 items adapted fromGowenet al.’s (1999) measure of leisure involvement, usinga 5-point scale ranging from1 (Strongly disagree) to 5(Strongly agree). Sample items included “I do theseactivities quite frequently” and “I devote a lot of mytime to these activities.” The coefficient a was .93.

We conducted supplementary analyses to supportthe positioning of job crafting and leisure activity asproactive behaviors—initiatives that people take toinfluence themselves and/or their environment(Bateman & Crant, 1993; Grant & Ashford, 2008).Following Hinkin and Tracey’s (1999) recommen-dations for assessing content validity, we surveyed148 individuals about the extent to which varioussurvey items—for job crafting, leisure activity, anda direct measure of proactive behavior (Sonnentag,2003)—reflected the definition of proactive behaviorprovided above (on a scale of 1 5 Extremely badmatch to 7 5 Extremely good match). The meanratings for job crafting and leisure activity itemswereon the high end of this scale (5.25 and 4.82, re-spectively). Indeed, the mean ratings for these be-haviors were in the same general position on theresponse scale as the means for the direct proactivebehavior items (5.54), though the difference wasstatistically significant. Moreover, when rated forfrequency on their typical Likert scales, job cratingand leisure activity weremoderately correlated withproactive behavior (.46 and .34, respectively). Com-bined, these results suggest that, although our mea-sures of job crafting and leisure activity are notredundant with proactive behavior, these behaviorscould be placed under a “proactivity” umbrella.

Job engagement. Participants rated their job en-gagement with Schaufeli and Bakker’s (2003) 9-itemmeasure, using a 5-point scale ranging from 1 (Al-most never) to 5 (Very often). Sample items included“At work, I feel bursting with energy” and “I amimmersed in my work.” The coefficient a was .93.

Task performance. Supervisors rated participants’task performance on a 5-point scale, ranging from1 (Unsatisfactory) to 5 (Outstanding), on the six di-mensions (e.g., “Dependability,” “Know-how andjudgment”) from Graen, Novak, and Sommerkamp(1982). The coefficient a was .92.

Citizenship behavior. Citizenship behavior wasmeasuredwith 8 items from Lee andAllen (2002) thatcapture citizenship behavior targeted at the organiza-tion. The5-point scale ranged from1 (Almost never) to5 (Very often). Example items included “Attends


functions that are not required but that help the com-pany’s image” and “Offers ideas to improve the func-tioning of the company.” The coefficient a was .93.

Analytical Strategy

The overall hypothesized model includes the in-direct effect of the incongruence of two variables(Individual’s values and Organization’s values)moderated by a third variable (either Job crafting orLeisure activity). Because there was not precedent forthis type of analysis in the literature, we derivedequations for conditional indirect effects of a congru-ence effect by integrating procedures for polynomialregression (Edwards & Parry, 1993) and moderatedmediation (Edwards & Lambert, 2007). These equa-tions, along with computations for evaluating theslope and curvature of the moderated direct and in-direct effects, are presented in Appendix A.

Our hypotheses—that negative effects of value in-congruencewill be tempered by job crafting or leisureactivity—suggest that the shape of the surface alongthe line of incongruence will be different at high ver-sus low levels of the moderators. To provide morecontext for interpreting our results, we have includeda figure that represents our expectations (seeFigure2).The line of incongruence (indicated by a dashed line)exists along the floor of the graph, from the pointwhere individual values are low and organizationalvalues are high to the point where individual valuesare high and organizational values are low.When themoderator is low, we expect that the slope of the sur-face along the line of incongruence will not be signif-icantly different than 0 (i.e., no differences betweenmisfits whose values are greater than those of the or-ganization and misfits whose values are less thanthose of the organization) and the curvature of thesurfacewill be significant andnegative (i.e., a concavesurface). In contrast, when the moderator is high, weexpect that both the slope and curvature of the surfacealong the lineof incongruencewill benot significantlydifferent than 0, indicating a flat surface.2

We followed other procedures outlined by re-searchers who have employed either the polynomialregression or moderated mediation methodologiesin a path analytic framework (e.g., Edwards & Cable,2009; Edwards & Lambert, 2007). Further, becauseproduct terms can produce Type I errors due to non-normal distributions (Shrout & Bolger, 2002), weproduced bias-corrected confidence intervals from1,000 bootstrapped estimates for all parameter esti-mates (Edwards & Lambert, 2007; Shrout & Bolger,2002). Finally, we investigated the influence of sev-eral control variables that might offer alternativeexplanations for the effects of our model. Our resultsdid not significantly change while controlling forrespondents’ perceptions of person–job fit, re-spondents’ proactive personality, respondents’ or-ganizational tenure, the industry in which therespondent worked, or a set of dummy-coded vari-ables representing the specific value cited as mostimportant by each respondent. Thus, following re-cent advice about the superfluous inclusion of con-trol variables (Becker, 2005; Carlson &Wu, 2012),wedid not include any of these control variables in ourfinal model.

RESULTS

The means, standard deviations, and zero-ordercorrelations among the variables are presented inTable 1. Because of strong correlations between themoderators and between the dependent variables, weconducted confirmatory factor analyses (CFA) sepa-rately on these measures. Results of the CFA on themoderators showed that the two-factor model inwhich job crafting and leisure activity were kept sep-arate was a better fit to the data (x2 (43)5 122.07, p,.001; comparative fit index (CFI) 5 .94; root meansquare error of approximation (RMSEA) 5 .10; stan-dardized root mean square residual (SRMR) 5 .037)than the one-factor model (x2 (44)5 359.34, p, .001;CFI 5 .76; RMSEA 5 .19, SRMR 5 .124; Dx2 (1) 5237.27,p, .001).Resultsof theCFAfor thedependentvariables showed that the two-factor model in whichtask performance and citizenship behavior were keptseparate was a better fit to the data (x2 (76) 5 272.26,p, .001; CFI5 .90; RMSEA5 .12; SRMR5 .042) thanthe one-factormodel (x2 (77)5 528.39,p, .001;CFI5.78; RMSEA5 .17, SRMR5 .08;Dx2 (1)5 256.13, p,.001). Further probing of these results using explor-atory structural equation modeling (see Marsh et al.,2009, for in-depth rationale for and explanation of thistechnique) indicated that model fit did not signifi-cantly improve through allowing individual items to

2 Our hypotheses did not involve effects for employeeswho have low congruence (i.e., where individual and or-ganizational values are both low) versus those who havehigh congruence (i.e., where individual andorganizationalvalues are both high); thus, we did not focus analyses onthe shape of the surface along the line of congruence.However, an anonymous reviewer pointed out that ourdata and analyses contained information that would allowus to examine congruence effects; these post-hoc analysesare presented at the end of the results section.


load onto multiple factors. Together, these results sup-port the distinctiveness of our measures.

Hypotheses 1 and 2 focused on the interactive ef-fects of value incongruence and job crafting. Table 2shows the path analytic results for the equations in-cluding job crafting in predicting job engagement,task performance, and citizenship behavior. Hypoth-esis 1 predicted that job craftingwould interactwithvalue incongruence in predicting job engagement,such that the negative effectswould beweakerwhenjob crafting was high. The significant F-statistic (F 53.44) of Step 2 of this equation indicates that job craft-ing significantly interactedwith value incongruence inpredicting job engagement.

The slope and curvature of the surface along theline of value incongruence at high and low levels ofthemoderator were computed by substituting valuesone standard deviation above and below themean ofjob crafting (Cohen,Cohen,West, &Aiken, 2003) into

Equation 5 (see Appendix A). As shown in Table 3and illustrated in Figure 3, at low job crafting, theslope of the surface along the line of value in-congruence was not significantly different than0 (qslope 5 .25, 90% CI [2.30, .74]); this result in-dicates that there is not a directional effect—there isno significant difference in engagement betweenmisfits whose personal values were greater than or-ganizational values compared to misfits whose per-sonal values were lesser than organizational values.The curvature of the surface along the line of valueincongruencewas significant andnegative (qcurvature52.39, 90% CI [2.70, 2.12]), indicating a concavesurface. The negative curvature suggests that jobengagement decreased as employee values divergedfrom organizational values. In contrast, at high jobcrafting, both the slope and curvature of the surfacealong the line of value incongruence were not sig-nificantly different than 0 (qslope 5 .10, 90% CI

TABLE 1Means, Standard Deviations, and Correlations among the Study Variablesa

Variables Mean SD 1 2 3 4 5 6 7

1. Individual’s values 4.37 0.71 –

2. Organization’s values 3.48 1.07 .16* –

3. Job crafting 3.60 0.76 .22* .34* –

4. Leisure activity 3.89 0.72 .23* .26* .40* –

5. Job engagement 3.50 0.78 .10 .39* .40* .42* –

6. Task performance 4.33 0.63 .32* .09 .24* .29* .32* –

7. Citizenship behavior 4.05 0.70 .27* .30* .30* .28* .48* .69* –

a n 5 193.*p, .05

FIGURE 2Hypothesized Surface Plotsa

a I5 Individual’s Values;O5Organization’s Values. The line of incongruence is depicted with the dotted line along the floor of the graph.


[21.06, 1.13]; qcurvature 5 2.12, 90% CI [2.67, .47]),indicating a flat surface. In other words, when jobcrafting was high, value incongruence did not sig-nificantly relate to job engagement.

Combined, this specific pattern of results supportsHypothesis 1. It suggests that misfits with lowerlevels of job crafting are more likely to suffer thenegative effects of value incongruence on job en-gagement, but misfits with higher levels of job craft-ingwere buffered against thenegative effects of valueincongruence on job engagement.

Hypothesis 2 predicted that the indirect effects ofvalue incongruence on (a) task performance and (b)citizenship behavior would be moderated by jobcrafting, such that thenegative indirect effectswouldbe weaker when job crafting was high. To determinewhether the effects of value incongruence carriedthrough to supervisor-rated job performance, theeffects of job engagement on task performance andcitizenship behavior were estimated while control-ling for the effects of the first stage of the model(i.e., the terms representing value incongruence, job

TABLE 2Path Analytic Results from the Estimated Model including Job Craftinga

Variables

Job Engagement

Step 1 Step 2 Task Performance CitizenshipBehavior

b SE b SE b SE b SE

Individual’s values (I) .14 .25 .22 .33 .40 .30 .14 .28Organization’s values (O) .01 .13 .04 .16 2.19 .14 .01 .17I2 2.08 .11 2.11 .14 2.09 .12 .02 .12I 3 O .14* .08 .09 .10 .05 .08 .02 .10O2 2.04 .04 2.05 .05 .11* .03 .07* .04Job crafting (JC) .31* .08 .26 .20 2.02 .20 .05 .16I 3 JC 2.26 .35 .09 .35 .32 .29O 3 JC 2.17 .23 2.08 .21 2.30 .23I2 3 JC .06 .17 2.03 .15 2.14 .14I 3 O 3 JC .07 .13 2.01 .11 .16 .13O2 3 JC .19* .07 .03 .04 2.06 .06Job engagement .24* .07 .39* .06F-statistic 3.44* 14.87* 34.30*ΔR2 .07* .06* .13*R2 .25* .31* .25* .32*

a n5193. Standarderrorsproduced from1,000bootstrappedestimates.F-statistic andΔR2 for taskperformance andcitizenshipbehavior arebased on the change in variance explained compared to the model excluding job engagement.

*p, .05, one-tailed

TABLE 3Tests of Response Surfaces along Line of Incongruence at High and Low Levels of Job Craftinga

Slope of Surface Curvature of Surface

Dependent Variable Level of Job Crafting qslope 90% CI qcurvature 90% CI

Job engagement Low .25 [2.30, .74] 2.39* [2.70, 2.12]High .10 [21.06, 1.13] 2.12 [2.67, .47]

Task performance Low .06 [2.06, .20] 2.10* [2.21, 2.03]High .03 [2.25, .28] 2.03 [2.18, .11]

Citizenship behavior Low .10 [2.11, .29] 2.15* [2.30, 2.05]High .04 [2.41, .43] 2.05 [2.27, .17]

a n 5 193. 90% bias-corrected confidence intervals produced from 1,000 bootstrapped estimates. qslope represents the slope of the surfacealong the line of incongruence (where individual values andorganizational values differ);qcurvature represents the curvature of the surface alongthe line of incongruence. Values for task performance and citizenship behavior represent the slope and curvature of the indirect effect of valueincongruence (via job engagement).

*p, .05, one-tailed


crafting, and their interaction). Referring back toTable 2, job engagement was seen to have significantpositive effects on task performance (b 5 .24) andcitizenship behavior (b5 .39). Following proceduresoutlined above, the slope and curvature of the in-direct effect of value incongruence on task perfor-mance along the line of incongruencewere computedat high and low levels of job crafting. As shown inTable 3 and illustrated in Figure 4a, when job craftingwas low, the surfaceof the indirect effect along the lineof incongruence was concave, as evidenced by a non-significant slope (qslope5 .06, 90%CI [2.06, .20]) anda significant and negative curvature (qcurvature52.10,90%CI [2.21,2.03]). This suggests that, as individualand organizational values diverged (i.e., as value in-congruence increased), task performance decreased.In contrast, when job crafting was high, the surface ofthe indirect effect was flat, as evidenced by a non-significant slope (qslope5 .03, 90%CI [2.25, .28]) andnon-significant curvature (qcurvature 5 2.03, 90% CI[2.18, .11]). This suggests that value incongruencedidnothave indirecteffects on taskperformancewhen jobcrafting was high.

Similarly, we evaluated the slope and curvature ofthe indirect effect of value incongruence on citizen-shipbehavior. Table3 andFigure4b show that,whenjob crafting was low, the indirect effect had a non-significant slope (qslope5 .10, 90%CI [2.11, .29]) andsignificant and negative curvature (qcurvature52.15,90% CI [2.30, 2.05]), which suggests a concavesurface along the line of incongruence. When jobcraftingwas high, both the slope (qslope5 .04, 90%CI[2.41, .43]) and curvature (qcurvature 5 2.05, 90% CI

[2.27, .17]) of the indirect effect were not significant,indicating a flat surface. Together, these resultsprovide full support forHypotheses 2(a) and 2(b) andsuggest that high job crafting buffers misfits againstthe negative indirect effects of value incongruenceon performance (via job engagement).

Hypotheses 3 and 4 focused on the interactiveeffects of value incongruence and leisure activity.Table 4 shows the path analytic results for theequations including leisure activity. Hypothesis 3predicted that leisure activity would interact withvalue incongruence in predicting job engagement,such that the negative effects would beweaker whenleisure activity was high. The significant F-statistic(F 5 2.39) of Step 2 of this equation indicates thatleisure activity significantly interacted with valueincongruence in predicting job engagement.

As shown inTable 5,when leisure activitywas low,the surface along the line of value incongruencewas consistent with our predictions. It had a non-significant slope (qslope 5 .31, 90% CI [2.20, .86]), in-dicating there were not significant differences inengagement between misfits whose personal valueswere greater than organizational values and misfitswhosepersonal valueswere lesser thanorganizationalvalues. The surface had a significant and negativecurvature (qcurvature 5 2.32, 90% CI [2.62, 2.07]),indicating a concave surface. Thus, as shown inFigure 5, below, when leisure activity was low, jobengagement decreased as individual values divergedfromorganizational values (i.e., asvalue incongruenceincreased). In contrast, when leisure activity was high,the negative effect of value incongruencewasmitigated.

FIGURE 3Relationship between Value Incongruence and Job Engagement at High and Low Levels of Job Craftinga

a I5 Individual’sValues;O5Organization’sValues; JE5 JobEngagement. The lineof incongruence isdepictedwith thedotted line along thefloor of the graph.


The shape of the surface along the line of incongruenceshowed a significant slope (qslope 5 2.90, 90% CI[22.09,2.00]) andnon-significantcurvature (qcurvature5.30, 90% CI [2.17, .87]. These results (shown inFigure 5) support Hypothesis 3 thatmisfits with lowerlevels of leisure activity are more likely to suffer thenegative effects of value incongruence on job engage-ment, but misfits with higher levels of leisure activitywere buffered against these negative effects. Althoughnot hypothesized, the pattern of results further sug-gests that leisure activity not only mitigates the nega-tive effect of value incongruence on job engagement,but also positively impacts job engagement for misfitswhose individual values are lower than the organiza-tion’s values.

Hypothesis 4 predicted that the indirect effectsof value incongruence on (a) task performance and(b) citizenshipbehaviorwouldbemoderatedby leisure

activity, such that the negative indirect effectwouldbe weaker when leisure activity was high. Referringback to Table 4, job engagement had significantpositive effects on task performance (b 5 .21) andcitizenship behavior (b5 .36). As shown in Table 5and illustrated in Figure 6a, when leisure activitywas low, the surface along the line of incongruencewas consistent with our predictions. The slope ofthe surface was not significantly different than0 (qslope 5 .07, 90% CI [2.02, .23], and its curvaturewas significant andnegative (i.e., concave;qcurvature52.07, 90% CI [2.16, 2.01]). When leisure activitywas high, the negative effect of value incongruencewas mitigated. The indirect effect along the line ofincongruence was sloped negatively (qslope 52.19,90% CI [2.55, 2.01]), decreasing from low indi-vidual values/high organization values to high in-dividual values/low organizational values, and its

FIGURE 4Indirect Effect of Value Incongruence on (a) Task Performance and (b) Citizenship Behavior (via Job

Engagement) at High and Low Levels of Job Craftinga

a I 5 Individual’s Values; O 5 Organization’s Values; TP 5 Task Performance; OCB 5 Organizational Citizenship Behavior. The line ofincongruence is depictedwith the dotted line along the floor of the graph. Graphs for low job crafting are depicted on the left side of the figure;high job crafting is represented on the right side.


curvaturewasnot significant (qcurvature5 .06, 90%CI[2.03, .22]). This suggests that high leisure activitybuffered the indirect effect of value incongruence ontask performance for misfits.

Similarly, we evaluated the slope and curvature ofthe indirect effect of value incongruence on citizen-ship behavior. Table 5 shows that, when leisure ac-tivity was low, the shape of the surface along the lineof incongruence was consistent with our pre-dictions: the slope of the indirect effect was not sig-nificant (qslope 5 .11, 90% CI [2.06, .34]) and the

curvature was significant and negative (qcurvature 52.11, 90% CI [2.24, 2.02]). Thus, as shown inFigure 6b, when leisure activitywas low, citizenshipbehavior decreased as individual and organizationalvalues diverged. In contrast, when leisure activitywas high, the effect of value incongruence wasmitigated. Similar to the effect for task performance,the slope of the indirect effect along the line of in-congruence was significant and negative (qslope 52.32, 90% CI [2.82, 2.01]) and the curvature wasnot significant (qcurvature 5 .11, 90% CI [2.06, .35]).

TABLE 4Path Analytic Results from the Estimated Model including Leisure Activitya

Variables

Job Engagement

Step 1 Step 2 Task Performance CitizenshipBehavior

b SE b SE b SE b SE

Individual’s values (I) .03 .27 2.03 .30 .19 .26 2.04 .22Organization’s values (O) .12 .15 .26 .16 2.10 .12 .08 .14I2 2.03 .12 2.00 .13 2.02 .11 .10 .10I 3 O .07 .08 2.03 .10 .01 .07 2.03 .09O2 2.03 .04 2.04 .05 .09* .03 .06 .04Leisure activity (LA) .36* .08 .48* .17 .31 .20 .10 .18I 3 LA 2.89* .35 2.40 .32 2.26 .33O 3 LA 2.05 .22 .06 .20 2.10 .25I2 3 LA .40* .16 .15 .14 .14 .16I 3 O 3 LA .06 .12 2.01 .11 .05 .14O2 3 LA .09 .06 2.02 .03 2.03 .04Job engagement .21* .07 .36* .06F-statistic 2.39* 11.02* 28.78*ΔR2 .05* .05* .11*R2 .27* .31* .27* .31*

a n5193. Standarderrorsproduced from1,000bootstrappedestimates.F-statistic andΔR2 for taskperformance andcitizenshipbehavior arebased on the change in variance explained compared to the model excluding job engagement.

*p, .05, one-tailed

TABLE 5Tests of Response Surfaces along Line of Incongruence at High and Low Levels of Leisure Activitya

Slope of Surface Curvature of Surface

Dependent Variable Level of Leisure Activity qslope 90% CI qcurvature 90% CI

Job engagement Low .31 [2.20, .86] 2.32* [2.62, 2.07]High 2.90* [22.09,2.00] .30 [2.17, .87]

Task performance Low .07 [2.02, .23] 2.07* [2.16, 2.01]High 2.19* [2.55, 2.01] .06 [2.03, .22]

Citizenship behavior Low .11 [2.06, .34] 2.11* [2.24, 2.02]High 2.32* [2.82, 2.01] .11 [2.06, .35]

a n 5 193. 90% bias-corrected confidence intervals produced from 1,000 bootstrapped estimates. qslope represents the slope of the surfacealong the line of incongruence (where individual values andorganizational values differ);qcurvature represents the curvature of the surface alongthe line of incongruence. Values for task performance and citizenship behavior represent the slope and curvature of the indirect effect of valueincongruence (via job engagement).

*p, .05, one-tailed


Together, these results provide support for Hypoth-eses 4(a) and 4(b) and suggest that high leisure ac-tivity buffers misfits against the negative indirecteffects of value incongruence on performance (viajob engagement).

Post-Hoc Analyses and Robustness Checks

We conducted analyses to examine our assump-tion that incongruence on an individual’s most im-portant value ismorepredictive of theoutcomes thanis incongruence on values ranked lower. First, weran our hypothesized model that simultaneouslyincluded terms for both the most importantand second-highest ranked value. The change inexplained variance was not significant when theterms for the second-highest ranked value wereadded. Further, the results for the most importantvalue were significant and consistent with those re-ported above.This pattern of results suggests that (in)congruence on themost important valuedoes indeedbest explain variance in the outcome variables.

Next, we examined several ways of including in-formation about the specific values (e.g., altruism,pay, prestige) in the analyses. For example, we rananalyses controlling for the specific value rankedhighest by each participant (coded into sevendummy-coded variables). The pattern of results wasessentially identical to the results we report above.Together, these additional results support the ideathat analyzing congruence on the top-ranked value

for each respondent is a reasonable and meaningfulapproach.

Although our research question is focused on thebuffering effects of job crafting and leisure activity formisfit (along the line of incongruence), our data andanalyses also allow for an investigation of the moder-ating effects of these variables along the line of con-gruence.3 In other words, we can examine whetherthese proactive behaviors might also benefit em-ployees with low value congruence (i.e., where indi-vidual values and organizational values are both low)compared to employees with high value congruence(i.e., where individual values and organizationalvalues are both high). The experience of low valuecongruence may leave core needs unsatisfied, com-pared to high value congruence (Edwards & Shipp,2007). Similar to the experience of misfit as we theo-rizedabove, jobcraftingand leisureactivitymaybufferthe performance consequences of low value congru-ence by supplementing for those unsatisfied needs.

We adapted the procedures listed in Appendix Ato investigate the shape of the surface along the lineof congruence. The adapted equations can be ob-tained by contacting the first author. Referring back toFigure 3, when job crafting was low, the slope of thesurface along the line of congruence was significantand positive (qslope 5 .59, 90% CI [.09, 1.27], and itscurvaturewas significant andnegative (i.e., concave;qcurvature52.32, 90%CI [2.66,2.04]). This suggests

FIGURE 5Relationship between Value Incongruence and Job Engagement at High and Low Levels of Leisure Activitya

a I5 Individual’sValues;O5Organization’sValues; JE5 JobEngagement. The lineof incongruence isdepictedwith thedotted line along thefloor of the graph.

3 We thank an anonymous reviewer for pointing this out.


that, when job crafting was low, job engagement in-creases at a decreasing rate as value congruence in-creases from low tohigh.When job craftingwas high,both the slope and curvature of the shape of thesurface along the line of congruence were not sig-nificant (qslope52.06, 90%CI [2.61, .62];qcurvature5.17, 90% CI [2.13, .44]), indicating that the surfacealong the line of congruence was flat. These effectsdid not carry through to task performance or citi-zenship behavior. Thus, it can be concluded that jobcrafting provides a buffer on job engagement forthose with low value congruence compared to thosewith high value congruence.

Referring back to Figure 5, when leisure activitywas low, the slope of the surface along the line ofcongruence was significant and positive (qslope5 .90,90% CI [.37, 1.51]), and its curvature was significantand negative (i.e., concave; qcurvature 5 2.47, 90% CI[2.81, 2.19]). This suggests that, when leisure activ-itywas low, job engagement increases at a decreasing

rate as value congruence increases from low to high.When leisure activity was high, the slope of thesurface along the line of congruence was not signif-icant (qslope 5 2.45, 90% CI [2.95, .36]) and its cur-vature was significant and positive (qcurvature 5 .33,90%CI [.01, .57]), indicating a convex surface. Uponexamination of the surface, job engagement wasgreater when value congruence is low or high thanwhen it is at the mean level. These effects did notcarry through to task performance but did to citi-zenship behavior. Thus, it can be concluded thatleisure activity provides a buffer on engagement andcitizenship behavior for those with low and highvalue congruence compared to those with meanlevels of value congruence.

DISCUSSION

The current economy has created an environ-ment where misfits are prevalent in the workplace.

FIGURE 6Indirect Effect of Value Incongruence on (a) Task Performance and (b) Citizenship Behavior (via Job

Engagement) at High and Low Levels of Leisure Activitya

a I 5 Individual’s Values; O 5 Organization’s Values; TP 5 Task Performance; OCB 5 Organizational Citizenship Behavior. The line ofincongruence is depicted with the dotted line along the floor of the graph. Graphs for low leisure activity are depicted on the left side of thefigure; high leisure activity is represented on the right side.


Unfortunately, a plethora of negative consequencesare associated with such value incongruence forthese employees, including lower satisfaction, en-gagement, and job performance (Edwards & Cable,2009; Kristof-Brown et al., 2005; Rich et al., 2010).From a company’s perspective, however, theremaybe unique benefits to retaining employees withdifferent values and perspectives (Harrison, 2007;Hoever et al., 2012; Schneider, 1987). In order torealize such benefits, it is critical to understandwhether and how misfits might be able to mitigatethese consequences so they can function moreproductively in a work environment that does notshare their important personal values (Kristof-Brown & Guay, 2011). Adopting the view that em-ployees can proactivelymanage their lives (Grant &Ashford, 2008), our research represents a step to-ward understanding the types of behaviors—suchas job crafting and leisure activity—that may miti-gate the negative consequences associated withvalue incongruence.

Theoretical and Practical Implications

At a global level, this study provides importantinsights into the work lives of misfits. The vast ma-jority of research in the value congruence literaturehas focused on one end of the fit–misfit spectrum—

on employees who fit well with their organizations(Kristof-Brown & Guay, 2011). Yet, the negativeconsequences associated with value incongruence(Kristof-Brown et al., 2005) and the prevalence ofmisfits in organizations todaymake it necessary thatwe also understand the experiences of misfits. Byintegrating motivational theorizing about need sat-isfaction to the value congruence literature, we pro-vide rationale for how misfits may compensate forvalue incongruenceandanswer thecall for increasedresearch attention on misfits (Judge, 2007; Kristof-Brown & Guay, 2011).

Going a step further, this study offers a host ofspecific theoretical contributions to the P–E fit andvalue congruence literatures. First, our results pro-vide novel information relevant to the dominantparadigm of the value congruence literature—ASA(Schneider, 1987). A key assumption of this theory isthat misfits are likely to leave their organizations insearch of a work environment in which they mightbetter fit. Yet, the reported relationship betweenvalue incongruence and turnover is not overly strong(Kristof-Brown et al., 2005), suggesting this as-sumption may be contingent on critical boundaryconditions. Indeed, our paper shows that misfits’

behaviors—specifically, their job crafting and lei-sure activity—can suppress the otherwise negativeeffects of value incongruence on their job engage-ment and performance. In this way, our researchhighlights the possibility that misfits can be pro-active “co-creators” of their work environment,whereby they combat the downsides of their situa-tion and create better work experiences for them-selves. Thus, our study adds nuance to Schneider’s(1987) famous proposition, suggesting that somepeople make their own place.

In integrating the idea of co-creation, this studyemphasizes the importance of individual agency inthevalue congruence equation andcalls for a broaderperspective in this area of P–E fit research. An im-portant goal of interactionist research is to “de-termine when and to what extent person andsituation variables predict behavior” (Chatman,1989: 333). With respect to individual and organi-zational values, our study suggests that it is alsocritical to ask for whom? Although value in-congruence has strong implications for the perfor-mance of some employees, others appear to haveinoculated themselves against its effects throughtheir own initiative. Indeed, not all employees can beconceptualized as “passive agents” (Chatman, 1989:337) who simply succumb to the work environmentas it currently exists. Yet, values are stable and en-during (Rokeach, 1973) and so it is difficult for anemployee to make a meaningful impact in thisregard. This article offers initial guidance on theindividual-level factors researchers should exploreto better understand how value congruence can af-fect employees.

This study further contributes to the P–E fit andvalue congruence literatures by incorporating therole of nonwork behaviors. Traditionally, these lit-eratures have focused narrowly on the role of workfactors to explain fit and its effect on outcomes(Edwards & Cable, 2009; Edwards & Rothbard, 1999;Erdogan, Kraimer, & Liden, 2004). By focusing onhow employees’ needs can be met through leisureactivity, this manuscript infuses a broader perspec-tive into the value congruence literature. Ultimately,this inclusion underscores the importance of ac-counting for cross-domain effects on the relationshipbetween value (in)congruence and its outcomes.

From a practical perspective, this manuscriptprovides important information for managers seek-ing to achieve a healthy balance between strongvalue congruence for individual employees andsufficient diversity inworkforce values. Notably, ourresults call into question strong recommendations


made in the popular press that hiring managers pri-oritize values above other factors (e.g., Margolis,2010; Rhoades, 2011). Chatman (1989: 344) speculatedthat “some optimal level of person–environmentfit may exist . . . in terms of the proportions of highand low ‘fitters’ within an organization.” That is,although value congruence generates importantbenefits for employees and organizations (Kristof-Brown et al., 2005; Schneider, 1987), a homogenousworkforce also presents significant drawbacks,such as increased rigidity, inability to adapt tochange, and stagnation (Harrison, 2007; Schneider,1987). Greater diversity on values also yields ben-efits, such as increased creativity (e.g., Hoever et al.,2012). Thus, understanding potential behavioralbuffers tovalue incongruence, suchas jobcraftingandleisure activity, may offer managers tools for helpingemployees manage incongruence on importantvalues, and ultimately finding the appropriateequilibrium in terms of high and low fitters. More-over, an individual’s tendency to engage in thesebehaviors may be an alternative—or, at the veryleast, complementary—consideration during theselection process.

Beyond the contributions to the P–E fit literature,this study also adds to conversations about jobcrafting and nonwork behavior. Job crafting is a newand growing literature that demonstrates a variety ofbehavioral gains, such as positive emotions, satis-faction, and job engagement (see Berg et al., 2013;Wrzesniewski et al., 2013). Ourmanuscript providesevidence that job crafting can also help peoplecompensate for significant pitfalls of their jobs—inother words, not only can job crafting make a good sit-uation better, but it may also helpmake a bad situationtolerable. Inaneconomywherevalue incongruencehasbecome more prevalent, this use of job crafting maybe particularly constructive (Wrzesniewski, Berg, &Dutton, 2010).

Although organizational research in the realm ofnonwork is also growing, the vast majority of thatliterature remains focused on the role of the familyand work–family interactions (Westring & Ryan,2010). Not surprisingly, work and family often con-flict (Greenhaus & Beutell, 1985), leading to theconclusion that nonwork pursuits detract from en-gagement and productivity at work. Our results—that leisure activity can compensate for a negativework experience—add to the growing evidence thatwork and nonwork pursuits can exist in harmony(e.g., Greenhaus & Powell, 2006).

Moreover, the specific pattern of this relationshipdemonstrated that leisure activity could go beyond

buffering the negative effects of misfit on perfor-mance to even improving the performance of somemisfits.We expected the relationship betweenmisfitand performance to be relatively flat when leisure washigh—indicating that misfit was no longer detrimentaltoperformance. Instead,wesawanunexpectedpositiveeffect for employees with a specific form of misfit—wherein their personal values were deemed less im-portant than the organization’s value (see Figures 5 and6). The fact that even their most important value is,relatively speaking, less important may indicate thatwork is simply less central to these individuals. That is,it is possible, instead, that the nonwork domain ismorecentral to these individuals—thus, they are morestrongly motivated by leisure activity. From a work/nonwork perspective, this finding lends credence tothe idea of enhancement effects between the two lifedomains (e.g., Greenhaus & Powell, 2006; Hakanen,Peeters, & Perhoniemi, 2011).

These findings elevate the importance of un-derstanding and incorporating leisure activity intoorganizational studies. Moreover, our focus on leisurepursuits also expands the nonwork discussion beyondthe traditional focusonfamily (Westring&Ryan,2010).As newer generations begin to join the workforce andbring with them a desire for life outside of work andpurpose in their activities (Twenge et al., 2010), re-search on nonwork pursuits, and leisure in particular,will be increasingly useful for organizations. Futureresearch may benefit not only from a continued focuson the role of leisure, but also from examining the roleof specific forms of leisure (e.g., exercise, entertain-ment, community involvement).

Limitations and Suggestions for Future Research

The nature by which we assessed fit—involvingparticipant ratings of both individual and organiza-tional values—introduces a potential for bias inreporting. In particular, it is not clear whether par-ticipants are the best source for rating organizationalvalues. It is possible that this approach does not trulycapture “objective fit.” However, we followed therecommended method for assessing value congru-ence, which is purported to be superior to the alter-native method of collecting direct reports ofperceived fit (Cable & Edwards, 2004; Edwards &Cable, 2009; Edwards, Cable, Williamson, Lambert,& Shipp, 2006).

Further, we tested hypotheses based on the valueranked as most important from the employee’s per-spective. This is a unique approach, as prior workhas examined congruence on each of eight values


separately (e.g., Cable&Edwards, 2004).Wepursuedthis direction because our focus was on mitigatingthe consequences of value incongruence, regardlessof the specific value at its root. Research on valuessuggests that people prioritize values differently andtheir actions are more influenced by more stronglyheld values (Edwards, 1992, 1996; Rokeach, 1973).An alternative route, however, would be to select thevalue rankedmost important to the company. Futureresearch may want to examine the experience ofmisfit based on that particular value. This approachmay allow scholars to examine new research ques-tions, such as whether employees are labeled byothers as misfits and the potential consequences ofsuch an experience.

An assumption inherent in our theorizing is thatvalue congruence, job crafting, and leisure activi-ties fulfill employee fundamental needs. Althoughprior theorizing and evidence suggests that theseexperiences are capable of providing for many ba-sic needs, such as meaning, belonging, and com-petence (Berg et al., 2010; Clary et al., 1998; Geroyet al., 2000; Greguras & Diefendorff, 2009), we donot directly test the role of need satisfaction in thisstudy. In future research, scholars may want toexplore the precise needs at the heart of these re-lationships, particularly the possibility that dif-ferent employee actions—such as job crafting andleisure activity—function by satisfying differentfundamental needs.

Conclusion

Recently, a variety of forces have contributed tolarger numbers of misfits existing in organizations.Thus, it is important for researchers, employees, andmanagers to better understand how misfits can mit-igate the negative and unfortunate consequencesassociated with value incongruence. This studyoffers one of the first examinations of proactivebehavior that helps provide a buffer against the mo-tivational and performance detriments of value in-congruence. The results show that both craftingone’s job to improve the experience of work and in-volvement in leisure activity can help misfits stayengaged and productive atwork.Managersmay helpemployees remain productive in their jobs by pro-viding opportunities for employees to craft their jobsin meaningful ways and suggesting greater in-volvement in fulfilling activities outside of workinghours—particularly for misfits, whose needs are notfully satisfied due to incongruence with organiza-tional values.

REFERENCES

Bateman,T.S.,&Crant, J.M.1993.Theproactivecomponentof organizational component:Ameasureandcorrelates.Journal of Organizational Behavior, 14: 103–118.

Baumeister, R. F. 2012. Need-to-belong theory. InP. A. M. Van Lange, A. W. Kruglanski & E. T. Higgins(Eds.), Handbook of theories of social psychology,vol. 2: 121–140. Thousand Oaks, CA: Sage.

Baumeister, R. F., & Leary, M. R. 1995. The need to belong:Desire for interpersonal attachments as a fundamentalhumanmotivation.PsychologicalBulletin,117:497–529.

Baumeister, R. F., & Vohs, K. D. 2005. The pursuit ofmeaningfulness in life. In C. R. Snyder & S. J. Lopez(Eds.), Handbook of positive psychology: 608–618.New York, NY: Oxford University Press.

Becker, T. E. 2005. Potential problems in the statisticalcontrol of variables in organizational research: Aqualitative analysis with recommendations. Organi-zational Research Methods, 8: 274–289.

Berg, J. M., Dutton, J. E., & Wrzesniewski, A. 2013. Jobcrafting and meaningful work. In B. J. Dik, Z. S. Byrne& M. F. Steger (Eds.), Purpose and meaning in theworkplace: 81–104. Washington, D.C.: AmericanPsychological Association.

Berg, J.M., Grant, A.M., & Johnson, V. 2010.When callingsare calling: Crafting work and leisure in pursuit ofunansweredoccupational callings.OrganizationScience,21: 973–994.

Borman, W. C., & Motowidlo, S. J. 1993. Expanding thecriterion domain to include elements of contextualperformance. In N. Schmitt & W. C. Borman (Eds.),Personnel selection in organizations: 71–98. SanFrancisco, CA: Jossey-Bass.

Burke, R. J., & Greenglass, E. 1987. Work and family. InC. L. Cooper & I. T. Robertson (Eds.), Internationalreview of industrial and organizational psychology:273–320. New York, NY: Wiley.

Cable, D. M., & Edwards, J. R. 2004. Complementary and sup-plementary fit: A theoretical and empirical integration.The Journal of Applied Psychology, 89: 822–834.

Carlson, K. D., & Wu, J. 2012. The illusion of statistical con-trol: Control variable practice in management research.Organizational Research Methods, 15: 413–435.

Chatman, J. A. 1989. Improving interactional organiza-tional research: A model of person–organization fit.Academy of Management Review, 14: 333–349.

Christian, M. S., Garza, A. S., & Slaughter, J. E. 2011. Workengagement: A quantitative review and test of its re-lations with task and contextual performance. Per-sonnel Psychology, 64: 89–136.

Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas,A.A.,Haugen, J., &Miene, P. 1998.Understanding and


assessing the motivations of volunteers: A functionalapproach. Journal of Personality and Social Psy-chology, 74: 1516–1530.

Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. 2003.Appliedmultiple regression/correlation analysis for the behav-ioral sciences (3rd ed.).Hillsdale,NJ: LawrenceErlbaum.

Colquitt, J.A.,Baer,M.D., Long,D.M.,&Halvorsen-Ganepola,M. D. K. 2014. Scale indicators of social exchange re-lationships: A comparison of relative content validity.The Journal of Applied Psychology, 99: 599–618.

Deci, E. L., & Ryan, R. M. 1985. Intrinsic motivation andself-determination in human behavior. New York,NY: Plenum.

Deci, E. L., & Ryan, R. M. 2000. The “what” and “why” ofgoal pursuits:Humanneeds and the self-determinationof behavior. Psychological Inquiry, 11: 227–268.

Edwards, J. R. 1992. A cybernetic theory of stress, coping,and well-being in organizations. Academy of Man-agement Review, 17: 238–274.

Edwards, J. R. 1996.Anexaminationof competing versionsof the person–environment fit approach to stress.Academy of Management Journal, 39: 292–339.

Edwards, J. R., & Cable, D. M. 2009. The value of valuecongruence. The Journal of Applied Psychology, 94:654–677.

Edwards, J. R., Cable, D. M., Williamson, I. O., Lambert,L. S., & Shipp, A. J. 2006. The phenomenology of fit:Linking the person and environment to the subjectiveexperience of person–environment fit. The Journal ofApplied Psychology, 91: 802–827.

Edwards, J. R., & Lambert, L. S. 2007. Methods for in-tegrating moderation and mediation: A general ana-lytical framework using moderated path analysis.Psychological Methods, 12: 1–22.

Edwards, J. R., & Parry, M. E. 1993. On the use of poly-nomial regressionas analternative todifference scoresin organizational research.AcademyofManagementJournal, 36: 1577–1613.

Edwards, J. R., & Rothbard, N. P. 1999. Work and familystress and well-being: An examination of person–environment fit in the work and family domains.Organizational Behavior and Human Decision Pro-cesses, 77: 85–129.

Edwards, J. R., &Rothbard,N. P. 2000.Mechanisms linkingwork and family: Clarifying the relationship betweenwork and family constructs. Academy of Manage-ment Review, 25: 178–199.

Edwards, J. R., & Shipp, A. J. 2007. The relationship be-tween person–environment fit and outcomes: An in-tegrative theoretical framework. In C. Ostroff &T. A. Judge (Eds.), Perspectives on organizational fit:209–258. San Francisco, CA: Jossey-Bass.

Erdogan, B., Kraimer, M. L., & Liden, R. C. 2004. Workvalue congruence and intrinsic career success: Thecompensatory roles of leader–member exchange andperceived organizational support. Personnel Psy-chology, 57: 305–332.

Geroy, G. D., Wright, P. C., & Jacoby, L. 2000. Towarda conceptual framework of employee volunteerism:Anaid for thehuman resourcemanager.ManagementDecision, 38: 280–286.

Gowen,M.A., Riordan, C.M., &Gatewood, R. D. 1999. Testof a model of coping with involuntary job loss fol-lowing a company closing. The Journal of AppliedPsychology, 84: 75–86.

Graen, G. B., Novak, M., & Sommerkamp, P. 1982. Theeffects of leader–member exchange and job design onproductivity and satisfaction: Testing a dual attach-ment model. Organizational Behavior and HumanPerformance, 30: 109–131.

Grant, A. M. 2012. Giving time, time after time: Work de-sign and sustained employee participation in corpo-rate volunteering.Academy ofManagement Review,37: 589–615.

Grant, A. M., & Ashford, S. J. 2008. The dynamics of pro-activity at work. Research in Organizational Be-havior, 28: 3–34.

Greenhaus, J. H., & Beutell, N. J. 1985. Sources of conflictbetween work and family roles. Academy of Man-agement Review, 10: 76–88.

Greenhaus, J. H., & Powell, G. N. 2006. When work andfamily are allies: A theory ofwork–family enrichment.Academy of Management Review, 31: 72–92.

Greguras, G. J., & Diefendorff, J. M. 2009. Different fitssatisfy different needs: Linking person–environmentfit to employee commitment and performance usingself-determination theory. The Journal of AppliedPsychology, 94: 465–477.

Hackman, J. R., & Oldham, G. R. 1980. Work redesign.Reading, MA: Addison-Wesley.

Hakanen, J. J., Peeters, M. C. W., & Perhoniemi, R. 2011.Enrichment processes and gain spirals at work and athome: A 3-year cross-lagged panel study. Journal ofOccupational and Organizational Psychology, 84:8–30.

Harding, R., &Mackenzie,M. 2014, January 10. U.S. jobs datapuncture recovery hopes.Financial Times. http://www.ft.com/cms/s/0/e9a9f980-79fc-11e3-a3e6-00144feabdc0.html. AccessedMarch 18, 2014.

Harrison, D.A. 2007. Pitching fits in applied psychologicalresearch: Making fit methods fit theory. In C. Ostroff &T. A. Judge (Eds.), Perspectives on organizational fit:389–416.NewYork,NY: LawrenceErlbaumAssociates.

Hecht, T. D., & Boies, K. 2009. Structure and correlates ofspillover from nonwork to work: An examination of


http://www.ft.com/cms/s/0/e9a9f980-79fc-11e3-a3e6-00144feabdc0.html



nonwork activities, well-being, and work outcomes.Journal of Occupational Health Psychology, 14:414–426.

Heine, S. J., Proulx, T., & Vohs, K. D. 2006. The meaningmaintenance model: On the coherence of social mo-tivations. Personality and Social Psychology Review,10: 88–110.

Hills, P., & Argyle, M. 1998. Positive moods derived fromleisure and their relationship to happiness and per-sonality.Personality and IndividualDifferences, 25:523–535.

Hinkin, T. R., & Tracey, J. B. 1999. An analysis of varianceapproach to content validation. Organizational Re-search Methods, 2: 175–186.

Hoever, I. J., van Knippenberg, D., van Ginkel, W. P., &Barkema, H. G. 2012. Fostering team creativity: Per-spective taking as key to unlocking diversity’s po-tential. The Journal of Applied Psychology, 97:982–996.

Judge, T. A. 2007. The future of person–organization fitresearch: Comments, observations, and a few sugges-tions. In C. Ostroff & T. A. Judge (Eds.), Perspectiveson organizational fit: 417–445. New York, NY:Lawrence Erlbaum Associates.

Judge, T. A., Simon, L. S., Hurst, C., & Kelley, K. 2013.“What I experienced yesterday is who I am today”:Relationship of work motivations and behaviors towithin-individual variation in the five-factormodel ofpersonality. The Journal of Applied Psychology, 99:199–221.

Kahn, W. A. 1990. Psychological conditions of personalengagement and disengagement at work.Academy ofManagement Journal, 33: 692–724.

Kahn, W. A. 1992. To be fully there: Psychological pres-ence at work. Human Relations, 45: 321–349.

Kristof, A. L. 1996. Person–organization fit: An integrativereview of its conceptualizations, measurement, andimplications. Personnel Psychology, 49: 1–49.

Kristof-Brown, A. L., & Guay, R. P. 2011. Person–environment fit. In S. Zedeck (Ed.), APA handbookof industrial and organizational psychology, vol. 3:3–50. Washington, D.C.: American PsychologicalAssociation.

Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C.2005. Consequences of individuals’ fit at work: Ameta-analysis of person–job, person–organization,person–group, and person–supervisor fit. PersonnelPsychology, 58: 281–342.

Latack, J. C., &Havlovic, S. J. 1992. Copingwith job stress: Aconceptual evaluation framework for copingmeasures.Journal of Organizational Behavior, 13: 479–508.

Lavelle, J. J., Rupp, D. E., & Brockner, J. 2007. Takinga multifoci approach to the study of justice, social

exchange, and citizenship behavior: The target simi-larity model. Journal of Management, 33: 841–866.

Leana, C., Appelbaum, E., & Shevchuk, I. 2009. Workprocess and quality of care in early childhood educa-tion: The role of job crafting. Academy of Manage-ment Journal, 52: 1169–1192.

Lee,K.,&Allen,N. J. 2002.Organizationalcitizenshipbehaviorandworkplace deviance: The role of affect and cognition.The Journal of Applied Psychology, 87: 131–142.

Light, J. 2010, May 26. More workers start to quit.Wall Street Journal. http://online.wsj.com/articles/SB10001424052748704113504575264432377146698.Accessed March 18, 2014.

Locke, E.A. 1976.Thenature andcauses of job satisfaction.In M. D. Dunnette (Ed.), Handbook of industrial andorganizational psychology: 1297–1350. Chicago, IL:Rand McNally.

Margolis, S. L. 2010. Building a culture of distinction.Atlanta, GA: Workplace Culture Institute.

Marsh, H. W., Muthen, B., Asparouhov, T., Ludtke, O.,Robitzsch, A., Morin, A. J. S., & Trautwein, U. 2009.Exploratory structural equation modeling, integratingCFA and EFA: Application to students’ evaluations ofuniversity teaching. Structural Equation Modeling,16: 439–476.

Meglino, B. M., & Ravlin, E. C. 1998. Individual values inorganizations: Concepts, controversies, and research.Journal of Management, 24: 351–389.

Meglino, B. M., Ravlin, E. C., & Adkins, C. L. 1989. A workvalues approach to corporate culture:A field test of thevalue congruence process and its relationship to in-dividual outcomes. The Journal of Applied Psychol-ogy, 74: 424–432.

Moore, B. 2010, May 10. The worst generation? New YorkPost. http://nypost.com/2010/05/10/the-worst-generation/.Accessed February 1, 2015.

O’Reilly, C. A., Chatman, J., & Caldwell, D. F. 1991. Peopleand organizational culture: A profile comparison ap-proach to assessingperson–organization fit.Academyof Management Journal, 34: 487–516.

Organ, D. W. 1988. Organizational citizenship behav-ior: The good soldier syndrome. Lexington, MA:Lexington Books.

Parker, S.K., Bindl,U.K., &Strauss,K. 2010.Making thingshappen: A model of proactive motivation. Journal ofManagement, 36: 827–856.

Rhoades, A. 2011. Built on values: Creating an enviableculture that outperforms the competition. NewYork,NY: Wiley.

Rich, L. R., LePine, J. A., & Crawford, E. R. 2010. Job en-gagement:Antecedents and effects on job performance.Academy of Management Journal, 53: 617–635.


http://online.wsj.com/articles/SB10001424052748704113504575264432377146698

http://online.wsj.com/articles/SB10001424052748704113504575264432377146698

http://nypost.com/2010/05/10/the-worst-generation/

Rodell, J. B., & Judge,T.A. 2009.Can “good” stressors spark“bad” behaviors? The mediating role of emotions inlinks of challenge and hindrance stressors with citi-zenship and counterproductive behaviors. The Jour-nal of Applied Psychology, 94: 1438–1451.

Rokeach, M. 1973. The nature of human values. NewYork, NY: Free Press.

Ryan, R. M., & Deci, E. L. 2000. Self-determination theoryand the facilitation of intrinsic motivation, social de-velopment, and well-being. The American Psychol-ogist, 55: 68–78.

Schaufeli, W. B., & Bakker, A. B. 2003. Test manual for theUtrecht Work Engagement scale. Unpublished man-uscript, Utrecht University, The Netherlands.

Schneider, B. 1987. The peoplemake the place. PersonnelPsychology, 40: 437–453.

Schwartz, S. H. 1992. Universals in the content andstructure of values: Theoretical advances and empir-ical tests in 20 countries. Advances in ExperimentalSocial Psychology, 25: 1–65.

Shrout, P. E., & Bolger, M. 2002. Mediation in experi-mental and nonexperimental studies: New proceduresand recommendations. Psychological Methods, 7:422–445.

Sonnentag, S. 2003. Recovery, work engagement, andproactive behavior: A new look at the interface be-tween nonwork and work. The Journal of AppliedPsychology, 88: 518–528.

Sonnentag, S.,Mojza, E. J., Demerouti, E., & Bakker, A. B.2012. Reciprocal relations between recovery andwork engagement: The moderating role of jobstressors. The Journal of Applied Psychology, 97:842–853.

Sthapit, A. 2010. Employee misfits: An emerging issuein HR acquisition. Touchstone (Nashville, Tenn.), 5:68–70.

Tinsley, H. E., & Eldredge, B. D. 1995. Psychological ben-efits of leisure participation: A taxonomy of leisureactivities based on their need-gratifying properties.Journal of Counseling Psychology, 42: 123–132.

Tinsley, H. E., Hinson, J. A., Tinsley, D. J., & Holt, M. S.1993. Attributes of leisure and work experiences.Journal of Counseling Psychology, 40: 447–455.

Trenberth, L., &Dewe,P. 2005.Anexplorationof the role ofleisure in coping with work related stress using se-quential tree analysis. British Journal of Guidance &Counselling, 33: 101–116.

Trougakos, J. P., Beal, D. J., Green, S. G., & Weiss, H. M.2008. Making the break count: An episodic examina-tion of recovery activities, emotional experiences, andpositive affect displays. Academy of ManagementJournal, 51: 131–146.

Twenge, J. M., Campbell, S. M., Hoffman, B. J., & Lance,C. E. 2010. Generational differences in work values:Leisure and extrinsic values increasing, social andintrinsic values decreasing. Journal of Management,36: 1117–1142.

U.S. Bureau of Labor Statistics. 2014. Job openings andlabor turnover [Press release]. http://www.bls.gov/news.release/pdf/jolts.pdf. Accessed March 31, 2014.

Westring, A. F., & Ryan, A. M. 2010. Personality andinter-role conflict and enrichment: Investigatingthe mediating role of support. Human Relations,63: 1815–1834.

Williams, K. D. 1997. Social ostracism. In R. M. Kowalski(Ed.), Aversive interpersonal behaviors: 133–170.New York, NY: Plenum Press.

Williams, L. J., & Anderson, S. E. 1991. Job satisfaction andorganizational commitment as predictors of organi-zational citizenship and in-role behaviors. Journal ofManagement, 17: 601–617.

Wrzesniewski, A., Berg, J.M., &Dutton, J. E. 2010. Turn thejob you have into the job youwant.Harvard BusinessReview, 88: 114–117.

Wrzesniewski, A., & Dutton, J. E. 2001. Crafting a job:Revisioning employees as active crafters of theirwork.Academy of Management Review, 26: 179–201.

Wrzesniewski, A., LoBuglio, N., Dutton, J. E., & Berg, J. M.2013. Job crafting and cultivating positive meaningand identity inwork. InA.B.Bakker (Ed.),Advances inpositive organizational psychology, vol. 1: 281–302.London, England: Emerald.

Ryan M. Vogel ([email protected]) is an assistant professorin the Department of Management at the Black School ofBusiness at Pennsylvania State University–Erie. He re-ceived his PhD from the University of Georgia’s TerryCollege of Business. His research interests includeemployee fit/misfit, value congruence, and abusivesupervision.

Jessica B. Rodell ([email protected]) is an associate pro-fessor in the Department of Management at the Universityof Georgia’s Terry College of Business. She received herPhD andMBA from the University of Florida’sWarringtonCollege of Business. Her research interests include em-ployee volunteering, justice, and emotions.

JohnW. Lynch ([email protected]) is an assistant professorin the Department of Managerial Studies at the Universityof Illinois at Chicago. His research interests include em-ployee volunteering, engagement, and identity manage-ment in the workplace.


http://www.bls.gov/news.release/pdf/jolts.pdf

http://www.bls.gov/news.release/pdf/jolts.pdf

mailto:[email protected]



APPENDIX A: EQUATIONS TO EVALUATE THEMODERATED-INDIRECT (MEDIATED) EFFECT

OF THE (IN)CONGRUENCE OF TWO VARIABLES

This integrated approach first involves the estimation ofthe mediator—job engagement—on the moderated effectsof value incongruence. The effects of value incongruenceon job engagement are represented by the equation:

M 5 a0 1 a1I 1 a2O1 a3I2 1 a4IO1 a5O2 1 e (1)

where M represents job engagement, I represents in-dividual values, and O represents organizationalvalues. The I2, IO, andO2 terms are included becauseestimating (in)congruence effects often requires non-linear and interactive terms (e.g., Cable & Edwards,2004; Edwards & Parry, 1993).

Before evaluating the interactive effects, the direct ef-fects of the moderator must be controlled for:

M 5 a0 1 a1I 1 a2O1 a3I2 1 a4IO1 a5O2 1 a6Z1 e

(2)

where Z represents the moderator, job crafting or lei-sure involvement.

Finally, the five terms representing the interactive ef-fects of value incongruence and themoderator are added tothe equation:

M 5 a0 1 a1I 1 a2O1 a3I2 1 a4IO1 a5O2 1 a6Z

1 a7IZ1 a8OZ1 a9I2Z1 a10IOZ1 a11O2Z1 e

(3)

where IZ, OZ, I2Z, IOZ, and O2Z terms collectivelyrepresent the interactive effects of value incongruenceand the moderator.

The interactive effects of the moderators are tested by ex-amining the incremental explained variance of Equation 3compared to Equation 2, as indicated by the F-statistic. If theincremental explained variance is significant, the coefficientsmaybeused further for hypothesis testing (seeEdwards, 1996).

Since the hypotheses involve the moderated effects ofvalue incongruence, we are primarily concerned with theshape of the surface along the line of incongruence (O52I),where important individual values diverge from organiza-tional values. The shape of the surface can be described intermsof its slopeandcurvature. Following recommendationsby Edwards and Parry (1993), these can be computed bysubstituting2I forO in Equation 3:

M 5 a0 1 a1I 2 a2I 1 a3I2 2 a4I2 1 a5I2 1 a6Z1 a7IZ

2 a8IZ1 a9I2Z2 a10I2Z1 a11I2Z1 e

(4)

After re-arranging and collecting like terms, the equa-tion becomes:

M 5 a0 1 ½a1 2 a2 1 ða7 2 a8ÞZ�I 1 a6Z

1 ½a3 2 a4 1 a5 1 ða9 2 a10 1 a11ÞZ�I2 1 e (5)

The slope of the surface along the line of in-congruence is represented by the quantity [qslope 5 a1 2a2 1 (a7 2 a8)Z]. Positive values for qslope indicatea slope that increases along the line of incongruence, fromthe point where individual values are low and organiza-tional values are high to the point where individual valuesarehigh andorganizational values are low.Negative valuesfor qslope indicate a slope that decreases along the line ofincongruence.Non-significant qslope values indicate a non-increasing and non-decreasing slope. The curvature of thesurface is represented by the quantity [qcurvature5 a32 a41a51 (a92 a101 a11)Z] (e.g., Edwards, 1996). Positive valuesfor qcurvature indicate the surface is convex (i.e., curved up-ward) along the line of incongruence. Negative values forqcurvature indicate the surface is concave (i.e., curveddownward) along the line of incongruence. Non-significantqcurvature values indicate a surface that is not curved. Theslope and curvature of the surface can be evaluated athigh and low levels of the moderator by substitutingvalues one standard deviation above and below themeanof the moderator, Z (Cohen et al., 2003). Using responsesurface methodology, the coefficients can be plotted tovisually examine the form of the relationship (Edwards &Parry, 1993).

Evaluating the moderated indirect effect of value in-congruence on the dependent variables, task performanceand citizenship behavior first involves estimating the di-rect effect of job engagement while controlling for themoderated-indirect effect of value incongruence on thedependent variables (Y):

Y 5 b0 1 b1I 1 b2O1 b3I2 1 b4IO1 b5O2 1 b6Z

1 b7IZ1 b8OZ1 b9I2Z1 b10IOZ1 b11O2Z

1 b12M 1 e

(6)

By substituting Equation 3 forM in Equation 6 and thenre-arranging and collecting like terms, the equation pre-dicting the dependent variable becomes:

Y 5 b0 1 a0b12 1 ðb6 1 a6b12ÞZ1 ½ðb1 1 a1b12Þ1 ðb7 1 a7b12ÞZ�I 1 ½b2 1 a2b12Þ1 ðb8 1 a8b12ÞZ�O1 ½ðb3 1 a3b12Þ1 ðb9 1 a9b12Þ�I2 1 ½ðb4 1 a4b12Þ1 ðb10 1 a10b12ÞZ�IO1 ½ðb5 1 a5b12Þ1 ðb11 1 a11b12ÞZ�O2 1 e (7)

To examine the curvature of the surface along the line ofincongruence for the indirect effect,2I is substituted forOin Equation 7. After re-arranging the equation and col-lecting like terms, Equation 8 is:


Y 5 b0 1 a0b12 1 ðb6 1 a6b12ÞZ1 ½½b1 2 b21 ðb7 2 b8ÞZ1 ½ða1 2 a2Þ1 ða7 2 a8ÞZ�ðb12Þ�I1 ½ðb3 2 b4 1 b5Þ1 ðb9 2 b10 1 b11ÞZ1 ½½a3 2 a4 1 a5 1 ða9 2 a10 1 a11ÞZ�ðb12Þ��I2 1 e

(8)

The slope of the curvature of the surface along the lineof incongruence includes the term representing the directeffect of value incongruence on the dependent variable(b1 2 b2), the term representing the direct effect of themoderator (b72 b8), and the product term representing the

moderated indirect effect of value incongruence [(a12 a2)1(a72 a8)Z][b12]. The curvature of the surface along the lineof incongruence includes the term representing the directeffect of job engagement (b3 2 b4 1 b5), the term repre-senting the direct effect of the moderator [(b9 2 b10 1 b11)Z], and the product term representing the moderated in-direct effect of value incongruence [a32 a41 a51 (a92 a101a11)Z][b12]. By substituting values one standard deviationabove and below the mean of the moderator, Z, into theproduct term, the indirect effect of value incongruence canbe computed, tested for significance, and plotted usingresponse surface methodology.


engaged and productive misfits: how job crafting...

Documents